Jayeetra Bhattacharjee

About

AI/ML Engineer (TCS) who built and deployed a production LLM-powered audit transaction validation service to reduce manual review of unstructured transaction records and comments. Implemented a LangChain/Python pipeline for extraction/normalization and discrepancy detection, with strong production reliability practices (decision logging, dashboards, labeled eval sets) and a human-in-the-loop auditor feedback loop to improve precision/recall under strict data-sensitivity and near-real-time constraints.

Experience

Education

Key Strengths

• Built and deployed a production LLM-based audit transaction validation system
• Strong handling of messy/unstructured data via preprocessing, validation, and schema-change resilience
• Designed extraction + normalization pipeline using regex and NLP/NER
• Implemented discrepancy detection using hybrid ML + rules and anomaly/outlier detection
• Established measurable evaluation with labeled historical data plus production logging and dashboards
• Closed-loop improvement using auditor feedback to tune prompts/rules and manage precision/recall tradeoffs
• Pragmatic model selection under constraints (data sensitivity, latency, precision), including self-hosted lightweight LLMs and structured JSON outputs
• Automated CI/CD release pipeline with GitLab CI (Docker build, unit tests, evals, staged deploys with approvals)
• Effective collaboration with non-technical stakeholders through demos, targeted questions, and business-metric-driven updates

Reference Highlights

• Excellent cross-team collaboration
• Effective with non-technical stakeholders
• Organized and proactive communicator
• Plans and prepares presentations in advance
• Delegates responsibilities to best-suited team members
• Contributed to production AI/LLM product deliveries
• Successfully shipped LLM-based product to production
• Strong performance collaborating across teams
• Effective at presenting working prototypes to cross-functional teams
• Able to explain complex technical concepts clearly
• Uses analogies to make technical topics understandable
• Focus on operational efficiency and automation